Estimation of significance levels and confidence intervals for first-order reversal curve distributions

Abstract

First-order reversal curve (FORC) distributions provide a means with which to describe a magnetic mineral assemblage in terms of coercivities and interaction fields. In recent years the use of experimentally derived FORC distributions has increased dramatically and they are being placed in an increasingly quantitative interpretational framework. An outstanding issue for calculation and interpretation of FORC data sets is the statistical significance that can be assigned to structures in experimentally determined distributions. Without this knowledge, the selection and characterization of structures that can be deemed interpretable within a FORC distribution is a subjective process. We demonstrate how FORC processing algorithms can be adapted to provide a measure of statistical significance and a confidence interval for each point in a FORC distribution. This information can guide measurement protocols and provides a more quantitative framework for interpretation of FORC distributions.